Comparison of the Saccadic Eye Movement Ability of Female Professional Basketball Players and Non-Athletes

Abstract

Athletes, particularly ball game players, benefit from focusing on fast-moving visual targets. In this study, we hypothesized that athletes, specifically professional basketball players, exhibit more accurate saccadic eye movements compared to non-athletes. To test this hypothesis, we assessed the eye movements of eight female professional basketball players from the Niigata Albirex BB (basketball player group) and compared them to eight sex-matched healthy volunteers from a previous study. Eye movements were measured using the Eye Score device. The visual target moved continuously for 20 s at a speed of 100 degrees/s in regular triangular wave-like movements horizontally. The subjects tracked the visual targets with maximum accuracy. The total amplitudes of evoked eye movements during tracking were 37.96 ± 0.82 for the basketball player group and 27.74 ± 2.55 deg (mean ± SEM) for the non-athlete group, indicating a significant difference. These findings suggest that basketball players may achieve accurate saccades by employing predictive saccades. The saccades of the basketball players were concentrated at the fifth frame (0.17 s after the onset of target motion), exhibiting a more peaked distribution than the normal curve. The results imply that athletes may execute predictive saccades when tracking targets with predictable movements, even outside of actual competition.

1. Introduction

Athletes, especially those engaged in ball games, must focus on fast-moving visual targets. Eye movements include involuntary actions like the optokinetic nystagmus, vestibular–ocular reflex, and smooth pursuit eye movements, as well as deliberate movements like vergence eye movements and saccadic eye movements (saccades) [1]. Among these, saccades are pivotal in quickly fixing the gaze on moving visual targets.

Research indicates differences in gaze utilization between non-athletes and athletes during sports play [2,3]. Land et al. suggested that skilled athletes, owing to their anticipatory abilities, could execute saccadic eye movements toward the anticipated position of the ball. However, this possibility remains unclear because reproducibility is limited in real gaming conditions. To address this, repeated measurements of eye positions under controlled laboratory conditions may be necessary. Conducting research in the laboratory offers advantages, allowing for the assessment of visual function in athletes through various ophthalmologic examinations and ensuring reproducibility—a key factor in cumulative research efforts.

One method for comparing eye movements between athletes and non-athletes is the dynamic visual acuity (DVA) test [4]. This test measures the ability to discern a horizontally moving visual target, typically represented by a Landolt “C” ring. Uchida et al. [5] conducted a study comparing the DVA of baseball players and non-athletes. They found no difference in the results when the subjects fixated on the front, but when the subjects tracked the target horizontally, the baseball players performed better than non-athletes. However, DVA measurements involve the detection of the orientation of the Landolt ring, making it challenging to isolate eye movement ability.

Clinical ophthalmology offers various eye movement measurement methods, including the invasive search coil method [6] and electro-oculography (EOG) [7]. The search coil method, which uses contact lens electrodes, is highly invasive, while EOG may introduce artifacts due to body movements and electrode misalignment, limiting its use in measuring athletes’ eye movements [8,9]. Video-based eye trackers have recently advanced, using a camera with infrared diodes to map reflection patterns on the subjects’ corneas, enabling eye movement measurement [10]. Although eye tracking is commonly employed to assess eye movement during sports play by synchronizing with external camera images [3], its predominant use has been during actual games. However, evaluating eye movements in a laboratory setting using an eye-tracking device requires adequate visual stimuli and a monitor for display before measurements. Due to the large size of these equipment, measurements are limited to specific laboratories. Moreover, each facility using its own stimuli makes it challenging to compare findings. An additional drawback is the time-consuming need for thorough calibration before measurements.

We engineered a comprehensive eye movement measurement device capable of presenting visual stimuli internally while concurrently capturing bilateral eye positions. This was achieved through modifications to an eye-tracking-based ophthalmic examination device (ORTe EYENAC, NAC Image Technology Inc., Tokyo, Japan) [11]. Measurements with this device revealed that some healthy people are good at eye movements while others are not [12].

In this study, we hypothesized that athletes can perform more accurate saccades compared to non-athletes. To examine this hypothesis, the eye movements of professional basketball players were measured and compared with those of healthy subjects as previously reported [12].

2. Materials and Methods

2.1. Participants

The subjects were eight female professional basketball players from the Niigata Albirex BB (basketball player group) with a mean age of 24.3 ± 2.4 years. Prior to analysis, it was verified that the subjects had a visual acuity better than 0.0 logMAR units, no strabismus, and stereoacuity measured by the Stereo Fly test (Stereo Optical Co., Inc., Chicago, IL, USA) better than 60 s. All data were deidentified before any analysis.

2.2. Methods

The Eye Score device, based on ORTe EYENAC, was employed to measure eye movements. The visual target utilized was a dot (0.775 deg) moving continuously for 20 s in regular triangular wave-like patterns horizontally. The subjects were instructed to track the visual targets as accurately as possible, with the movement speeds set at 20, 40, 60, 80, and 100 deg/s. For the non-athlete group, 8 sex-matched volunteers were randomly selected from the normal volunteers reported in a previous study [12].

In the earlier study [12], gaze was recorded 3–5 frames (mean, 4.5) after the visual target. To address the time difference between eye movements and visual target recordings, the firmware was updated to compensate for this lag. This enhancement allowed for accurate evaluation of the time lag between the visual target and gaze. Consequently, a reanalysis of previously reported data for normal subjects was conducted to determine the time lag between the visual target and gaze, as well as the timing of saccades in normal subjects. In this experiment, both the basketball player group and the non-athlete group were measured for eye movements without any practice beforehand. All measurements took less than 5 min per subject, and no time was allowed for practice during the test.

For one of the eight participants, the right eye detection failed multiple times due to irregular reflections from the scleral contact lens worn exclusively on the right eye. Consequently, the subsequent analysis focused on left eye movements. The cross-correlation between the horizontal movements of both eyes was highly correlated (0.993 to 0.999; mean 0.997; Pearson’s product–moment correlations at the target velocity of 100 deg/s) with a zero lag-time, consistent with findings from our previous study [12].

2.3. Analysis

Similar to the previous study, we calculated cross-correlation coefficients for 575–600 sample pairs between the left eye position data with concurrently logged positions of the visual target. This analysis aimed to evaluate the precision of gaze tracking toward the visual target. To accommodate the temporal lag between visual target movements and eye responses, we determined the maximum correlation (rho max) within a time window spanning ±25 frames. To determine if gaze reached the target during saccades, the gaze amplitude was examined at a target speed of 100 deg/s. The total amplitude, calculated as the difference between the maximum and minimum eye positions for each cycle (looking right and left) was assessed. Since the target moved 20° in each direction, accurate saccades would yield a total amplitude of 40°. The F test was employed to assess differences in the variance of total amplitude between the basketball player group and the non-athlete group. Additionally, the total amplitudes of evoked eye movements in the two groups were compared using Welch’s two-sample t-test.

Eye movements with velocities exceeding 100 deg/s were classified as saccades [13,14]. The timing of saccades was based on the time lag at the onset of saccades. Histograms depicting the timing and frequency of saccades were generated, and kurtosis and normality were assessed using Geary’s test for normality concerning kurtosis.

2.4. Ethics Approval and Consent to Participate

This study was conducted with the approval of the Niigata University of Health and Welfare committee (18135-190117). The experiment was conducted in accordance with the Declaration of Helsinki, and written informed consent was obtained from all subjects.

3. Results

Figure 1 shows the typical eye movements evoked by a repeated motion of a visual target in a basketball player across the five target speeds. In good agreement with our previous findings [12], the participants could visually track the target even at 100 deg/s. Although the amplitudes of eye movements decreased with increasing target speed, they maintained the same repetitive motion pattern as the visual target, with only a minor bilateral deviation due to convergence eye movements.

To compare the eye movements of basketball players and non-athlete students, we calculated the rho max between the eye movements and target trajectories for both groups [12] (Figure 2).

At the highest visual target speed (100 deg/s), the horizontal eye movements mainly consisted of saccades (Figure 3). The visual target traversed from 20° to the left to 20° to the right; thus, the total amplitude would be 40° if exact saccades occurred. Figure 4 displays the total amplitude for each cycle of the basketball group and non-athlete group, with a mean total amplitude of 37.96 ± 0.82 and 27.74 ± 2.55 deg, respectively, which were significantly different based on Welch’s two-sample t-test (p = 0.0047) (Figure 5). There was also a significant difference in the variance of the trial-wise variances across the eight subjects (F test, p = 0.0020). This suggests that compared to the non-athlete subjects, basketball players executed more accurate and consistent eye movements in response to the regularly repeating movements of the visual target.

The time lag and frequency of saccades are presented in Figure 6, while Figure 7 depicts the integrated results of saccades during rightward and leftward gazes shown in Figure 6. Kurtosis was calculated for the time lag of saccade occurrence. At a target speed of 100 deg/s, the saccades of the basketball player group were concentrated at specific timings, exhibited high kurtosis, and displayed a non-normal distribution (kurtosis = 8.50, p = 0.0050). In contrast, in the non-athlete group, saccades were not concentrated at specific timings, showed low kurtosis, and were normally distributed (kurtosis = 4.11, p = 0.2110). Concerning the variance of lag during saccades, the variance in the basketball group was significantly smaller (F test, p = 0.0499). This result indicates that in the basketball player group, the timing of saccades is more concentrated than in the non-athlete group.

4. Discussion

To compare the saccade ability of basketball players with that of nonathletes, we assessed their capability to track a target moving 20° left and right (total 40°) at 100 deg/s. The total amplitude during tracking was 37.96 ± 0.82 deg for the basketball player group, significantly larger than 27.74 ± 2.55 deg for the non-athlete group. The saccades of the basketball player group were also concentrated at the fifth frame (0.17 s after the onset of target motion), exhibiting a sharper peak compared to the normal distribution. In contrast, the distribution of saccade timings in the non-athlete group followed a normal distribution. The variance of lag during saccades in the basketball group was significantly smaller. In the basketball player group, the timing of saccades was more concentrated, and the position of gaze during saccades was more accurate than in the non-athlete group. The basketball players were able to execute saccades to the appropriate position for the visual target with predictable movements, even though they were performing the task for the first time.

In a situation simulating a game, expert volleyball players exhibit a prolonged ability to track the ball compared to near-expert players [15]. However, in real-game simulations, experienced players might utilize various cues to predict ball movement. For instance, Damian et al. [16] conducted an experiment using a temporal occlusion paradigm, presenting a video of a tennis player to both expert and novice tennis players. The experts demonstrated the capability to predict the ball’s course even when the video was occluded just before the opponent’s racket touched the ball during a serving motion. To evaluate an athlete’s eye movement ability per se, it is essential to measure it under conditions without influences from these complex cues. In this study, the accuracy of saccades between athletes and non-athletes was compared using a task involving tracking a target with predictable, regular movements. The results suggest that the saccades were accurate because athletes excel at predicting target movement.

Land et al. [2] reported that skilled cricket batsmen make predictive saccades during play. In this study, the latencies of saccades were four to five frames (0.13 to 0.17 s) in both groups, slightly shorter than the standard one (0.2 s) [17], indicating the involvements of anticipatory saccades even in the non-athletes. However, the saccades of the basketball player group were concentrated at the fifth frames (0.17 s) after the target started moving. Our findings indicate that athletes may execute predictive saccades against targets with predictable movements more regularly, even outside of actual competition.

The relationship between sports and visual function remains unclear, as Nascimento et al. [18] pointed out that sports vision is a relatively new specialty, requiring more scientific evidence to confirm the benefits of visual training in the sports field. Although it has been reported that training improves performance in simple reaction tasks [19,20], the extent to which this practice effect affects sport play is unknown [21]. Our results suggest that measuring athletes’ visual functions using this device in the future will be valuable for advancing the research relationship between sports and vision.

The limitation of this study is the low sampling rate (30 Hz) of the device, which makes detailed analyses of the dynamics of saccades difficult. Additionally, eye movement speeds exceeding 100 deg/s were defined as saccades based on previous studies, but it is possible that saccades were also included in eye movement speeds less than 100 deg/s. Also, because of the current limitations of our device, we only assessed fronto-parallel eye movements in this study. Since saccade and vergence eye movements are linked [22], it is important to evaluate vergence in order to examine the eye movement ability of athletes. We are planning to update our device to be capable of two different images for each display in order to create binocular disparities. With the improved device, more prominent differences will be found between athletes and non-athletes. The small sample size is another limitation. However, given the significant difference observed between the two groups, the sample size was deemed sufficient. Therefore, the possibility of a type 1 error occurring is considered to be extremely low.

5. Conclusions

We assessed the saccades of female professional basketball players utilizing the eye movement evaluation device “Eye Score” that we developed. The findings revealed that the basketball player group could bring their gaze closer to the target compared to the non-athlete group. Additionally, the basketball players exhibited a saccade latency of 0.17 s, shorter than the typical saccade latency of 0.2 s. These results imply that basketball players may employ predictive saccades, enabling them to precisely track rapidly moving visual targets.